Metraipolis › archive › dc_763f2e18274b

Level-of-Token Diffusion — arXiv:2610.05816

Source
arXiv
Original
https://arxiv.org/abs/2610.05816
License (as archived)
CC0-1.0
Archived
2026-10-06T05:27:11.205Z
SHA-256 of the archived text
3e19ebbaa14af511df793d74c3576db938dd8709e96f8eed12df9e1e186810b3
Stable URL of this page
https://metraipolis.com/documento/dc_763f2e18274b

How to cite

Level-of-Token Diffusion — arXiv:2610.05816 — arXiv, https://arxiv.org/abs/2610.05816 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_763f2e18274b (Metraipolis archive dc_763f2e18274b, sha256 3e19ebbaa14af511df793d74c3576db938dd8709e96f8eed12df9e1e186810b3)
BibTeX
@misc{dc763f2e18274b,
  title = {Level-of-Token Diffusion — arXiv:2610.05816},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.05816},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_763f2e18274b; sha256 3e19ebbaa14af511df793d74c3576db938dd8709e96f8eed12df9e1e186810b3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.05816. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Image and video diffusion models allocate equal computation to every region, even when the intended scene calls for varying levels of detail. The spatial distribution of detail can often be anticipated before generation, indicating where computation can be reduced. We introduce Level-of-Token (LoT) Diffusion, a framework that turns this knowledge into an explicit multiresolution token layout (Level-of-Token layout) for adaptive and efficient generation. Tokens represent rectangular patches of varying sizes and shapes, allocating finer tokens where detail is needed and coarser tokens elsewhere. We adapt pretrained diffusion transformers to LoT layouts through a patch-wise asymmetric flow parametrization and embeddings for multiresolution tokens, preserving full-resolution flow prediction at every denoising step while processing only a reduced token sequence. LoT Diffusion enables layout-adaptive generation while preserving pretrained generative priors. We demonstrate LoT with layouts derived from semantic masks, bounding boxes, texture variance, and depth-of-field cues, as well as agentic plans. Across image and video generation, LoT offers favorable quality-efficiency tradeoffs, with significant speedups determined by the layout's token budget. Our project website is at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.05816 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-06T05:24:06.073Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).